Cell motility plays a central role in tissue morphogenesis, embryonic development, angiogenesis, and wound healing. The primary techniques for inducing directional cell migration in vitro, for bottom-up assembly of tissue patterns, wound healing, and cell- on-chip devices, rely on external gradients, chemotaxis. However, due to the finite range of gradients that cells respond to, steering the migration of large population of cells independently over long distances remains a challenge. Over the past funding period, we reported innovative techniques for guiding cell migration using: 2D surface micropatterns (MANDIP - Microarray Amplification of Natural Directional Persistence) and 3D microchannels (TANDIP - Topographical Amplification of Natural Directional Persistence). Free from gradients, MANDIP and TANDIP are highly scalable and can operate in parallel to direct large number of cells over unlimited distances on complex paths. Demonstrated applications of this capability include MANDIP directed self- assembly of different cell populations into spatially defined structures and sorting of cells by their intrinsic motility. Using MANDIP, TANDIP, and a new technique for fabricating substrates with light-switchable cell adhesiveness, we can now probe mechanistically how cells sense their surrounding extracellular matrix (ECM) to polarize and migrate directionally either along or against their initial morphological polarization. We also partnered with 4 collaborators at UC Medical School and Childrens Hospital who specialize in migration signaling pathways and cell polarity. In this first competitive renewal application, we will focus on understanding the sequence and interdependency of key events that occur during cell polarization and migration by addressing the following specific questions: Aim 1. How do local changes in attachment of lamellipodia alter morphological polarization? Aim 2. How do geometric cell-substrate interactions direct Golgi polarization? Aim 3. How do different cell types differ in their response to biomechanical cues?